Integration of multi-time-scale models in time series forecasting

A solution to the problem of producing long-range forecasts on a short samp ling interval is proposed. It involves the incorporation of information fro m a long sampling interval series, which could come from an independent sou rce, into forecasts produced by a state-space model based on a short sampli ng interval. The solution is motivated by the desire to incorporate yearly electricity consumption information into weekly electricity consumption for ecasts. The weekly electricity consumption forecasts are produced by a stat e-space structural time series model. It is shown that the forecasts produc ed by the forecasting model based on weekly data can be improved by the inc orporation of longer-time-scale information, particularly when the forecast horizon is increased from 1 year to 3 years. A further example is used to demonstrate the approach, where yearly UK primary fuel consumption informat ion is incorporated into quarterly fuel consumption forecasts.